1-D phenomenological modelling of gas-solid fluidized bed reactors for chemical looping reforming
MetadataVis full innførsel
This PhD thesis is a part of the EU FP7 Project called NanoSim. The main objective of this work is to develop a 1-D phenomenological model for fluidized bed reactors to simulate the performance of second-generation CO2 capture and storage processes (technologies ready for commercialization by the year 2030 and beyond). The main purpose of using a 1-D phenomenological model instead of a more complex fundamental formulation is to provide valuable and sufficiently accurate information of industrial interest with less computational costs. The 1-D phenomenological model consists of a generic formulation based on an averaging probabilistic approach used to interface between the models (closure laws) for different fluidization regimes. Hence, it can be used under bubbling, turbulent and fast fluidization regimes. The developed model was used to investigate the performance of three different chemical looping reforming technologies: the conventional chemical looping reforming (CLR), the novel gas switching reforming (GSR) and the fuel reactor of the membrane assisted chemical looping reforming technology (MA-CLR). Moreover, it was also linked with models at the plant scale and used to study the effect of changes in the CLR process on the techno-economic performance of the natural gas (NG) fired power plant with pre-combustion CO2 capture with chemical looping reforming process. The performance of the conventional CLR process was compared against the novel GSR technology. This analysis suggested that the CLR process is best suited to thermal power production with pre-combustion CO2 capture, while the GSR process is best suited to pure hydrogen production. The simulations of the process also indicated that the solids circulation rate has a positive effect on the methane conversion; however, there is need to strike a balance between the circulation of solids and the complexity of the system or mixture of the gases for CLR and GSR, respectively. Most published models for fluidized be reactors assume isothermal conditions due to the excellent heat transfer properties within fluidized beds. In this thesis, the effect of the axial temperature gradients on the reactor performance was investigated for the MA-CLR fuel reactor. It was observed that the development of these axial temperature gradients inside the reactor can enforce a ~10% lower gas throughput than predicted by isothermal reactor modelling. For the techno-economic assessment of the NG fired power plant with precombustion CO2 capture with CLR process, a sensitivity study was carried out. In this study, the air flowrate to the air reactor, air reactor outlet temperature and the steam flowrate to the fuel reactor of CLR were changed and their effect on the techno-economic performance of the process was investigated. The results indicated that it is necessary to seek an optimum between the net electrical efficiency and the CO2 avoidance, which depend on the CH4 conversion in the CLR and on the conditions for water gas shift reaction. Furthermore, the levelised cost of electricity (LCOE) of the process was also estimated. It was found that it is very sensitive to the fuel costs and process contingency costs varying between 75.3 $/MWh and 144.8$/MWh. This study also helped identifying trends that help in deciding appropriate process design conditions. The use of empirical closure laws to describe the hydrodynamics within phenomenological models limits its range of applicability unless they are developed for the system under study. Thus, the development of new closures from more complex simulations (computational fluid dynamics simulations) is highly recommended for further work in order improve the model accuracy.